Highlights on the Study of Progress, Part 2
Interesting New Ideas from the Roots of Progress Fellows
Below are selections from a few more of the fellows.
The Free Market Futurist is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
I’ll take them three at a time and link you to their newsletters (mostly on Substack), so please give them a follow.
Paige Lambermont usually writes about energy for the Competitive Enterprise Institute, and check her out there as she points out the absurd regulations we keep putting on energy production. But the most charming piece Paige wrote was a personal observation about the pleasures of keeping physical books in this digital age.
Whenever I’m in an Airbnb, or a house staged for sale, or any liminal space that bookshelves have been inserted into as a decorative flair, I find myself scanning the spines, looking for favorites, and guessing if it’s a lovingly curated collection, or books sold by the foot for decor purposes based on nothing but cover color.
Looking at bookshelves in all of these spaces is entertaining, but there’s nowhere I enjoy it more than at a friend’s house. I’ve long joked with some of my friends that I would have known we’d be friends by looking at their bookshelves alone.
One of my oldest friends gave me a perfect glimpse into this recently. We were at her house. She had recently moved and hadn’t arranged her new bookshelves yet. We were drinking coffee in her kitchen—one of those far between aimless days with a friend that you get fewer and fewer of as adult life piles up. She knows how I am about books, so she asked if I’d like to help her arrange her new bookshelves. Instant yes. We moved the stacks of books from one room to another, chatting and picking up our coffees at intervals as we carried stacks between the rooms, shelving, categorizing, and deciding what belonged where.
We’ve been friends since junior high, and many of the books we’ve read have been things we came into together. That whole day we chatted about the books as we shelved them. The ones that we’d both read and could remember passing back and forth. The same golden and silver age science fiction picked up in early college, the Phillip K. Dick and Haruki Murakami books we picked up in high school. The books read for classes we’d taken together, or told each other about on the phone later when we were away at separate colleges. The philosophy books, and the classics we had in common. As we got older we’d send each other our own favorites as birthday presents, as if to say “read this piece of me you don’t yet know.” I sent her Brandon Sanderson, Douglas Adams, my favorite Agatha Christie. She sent me Frederik Pohl and Kim Stanley Robinson. Our taste in books is inextricably linked. We learned to love these things together. I knew all of this, but going through her shelves that day really reminded me just how deep that connection was. Reading the same books, and at the same unique stages of life will connect people in ways that few things can.
And yes, books sold by the color are a real thing.
Also, please be advised that if I ever visit your house, I will be judging you on your book collection.
Malcolm Cochran wrote a very interesting pair of pieces on land-use statistics. No, really.
He arrives at a very counter-intuitive conclusion, which is that industrialization and fossil fuels actually free up more land and more natural spaces, because they are far less extensive than agricultural uses. In his first piece, he describes how the automobile saved the forests of New England. In a follow-up, he concludes, with apologies to Joni Mitchell, that “We Didn't Pave Paradise, We Plowed It.”
[A]griculture takes up huge tracts of land—the equivalent of the Americas plus a large part of Asia—while our cities, towns, and roads (and anything else classified as an “artificial surface” by the FAO) have a footprint the size of Libya—just one percent of global land area.
These figures surprised me when I first saw them since I, like most people in wealthy industrial nations, spend almost all my time within that small purple zone. Living most of our lives within built-up areas screws up our perception of the environment. Since we’re rarely ever more than a few feet from a road or parking lot, we might assume, for example, that “we paved paradise” is an apt metaphor for what’s happened to the planet over the last few centuries. The reality is closer to “we plowed paradise.”
While our total energy consumption has massively increased since the Industrial Revolution, almost all of that growth was supplied by fossil fuels, which, since they are highly energy-dense and come from underground, don’t take up much land. Agriculture creates a different form of energy: food, which is very land-intensive (recall that we use over a third of the Earth’s land for farming). Humans need food. And historically, we needed horses and other beasts of burden, which also need food. Now we approach a general principle: when we replace things that eat with things that don’t, we save land. That is also why one of the best things you can do for the environment is to eat less meat.
So, Toby, we should have a blinkered focus on energy production because the vast majority of humanity’s land footprint comes from producing energy in the form of food.
Things don’t always turn out the way you expect. Some Modernist architects assumed that the rise of the industrial era meant that we would all be sitting on mass-produced plastic furniture in glass and steel apartment towers—mass-produced industrial materials for an age of industrial mass-production. What has mostly happened is that we place more value than ever on items that are hand-crafted out of natural materials. (Go browse the Thomas Moser catalogue some time.) As we become wealthier, we are able to afford that which is not industrial and mass-produced.
One of the counter-intuitive conclusions I draw from Malcolm’s piece is that a wealthy, hyper-technological, energy-abundant future will be one with more natural space, not less. I write this from my dentist’s office, with a magnificently bucolic view overlooking the Western slope of Monticello and part of Charlottesville, a heavily wooded town nestled in among the foothills of the Blue Ridge Mountains. This is representative of what the future of human civilization will look like.
Follow Malcolm at.
Some of the most interesting pieces written by the progress fellows were about how to achieve progress, and particularly about the methods of scientific discover and technological innovation. MIT physicist Florian Metzler co-wrote a very interesting piece on “intellectual dark matter,” ideas that exist in a kind of limbo between the crackpot theory and the proven breakthrough. His main example is the invention of the transistor.
Typically associated with two “magic months” in late 1947 and three “brilliant inventors” the commonly taught history of the transistor has been greatly simplified and romanticized. A closer look reveals that solid-state amplification effects—the basis for transistor operation—were already reported in published articles as early as the 1920s. However, replication attempts frequently failed and even those who claimed to see amplification effects could obtain them only erratically.
There were vastly different responses to such reports. Some, like Radio News editor Hugo Gernback, proclaimed already in 1925 a coming new age of miniaturized solid-state switches. Others, like the luminary physicist Wolfgang Pauli, discouraged his mentees to work in this area for its lack of credibility and reproducibility. In a letter to his junior team member Rudolf Peierls, Pauli called semiconductor research “Schmutzphysik”—the physics of dirt—and warned: “one shouldn’t work on semiconductors, that is a filthy mess; who knows whether semiconductors even exist.”
Yet the most constructive response came from those who were able to neither entirely accept nor preemptively dismiss these early anomaly reports: people like Mervin Kelly, research director at Bell Labs, who in the mid 1930s started to assemble a team of physicists, materials scientists, and tinkerers to carefully look into the claims that had been circulating in various technical communities he had tapped into.
Kelly struck a balance between going all in on an idea whose validity was not established and letting it lay by the wayside. Instead, he and his colleagues gradually built up a repository of expertise across knowledge areas that included surface physics, electron band theory, and high- quality single-crystal growth, the importance of which was initially dismissed by members of Kelly’s team like the physicist William Shockley.
Florian concludes that we need to devote more effort to chasing down uncertain and anomalous early results.
What lessons can we learn from such historical examples? Rather than watching out for Eureka moments, it would be wise to embrace those earlier more enigmatic leads that indicate that something interesting might be happening in a particular experimental configuration. Bardeen and Brattain’s polished 1948 Physical Reviewpaper on transistor action as well as Hahn and Strassmann’s polished 1938Naturwissenschaften paper on fission depended on these earlier phases and represented the culmination of many insights gained previously. At the same time these papers are effectively stripped and sanitized of the messy prehistories that enabled them in the first place. Rather than looking out for the next Bardeen and Brattain transistor paper or the next Hahn and Strassmann fission paper, scientific institutions today would do well to seek out the equivalents of those 1925 anomalous amplification reports and of those 1934 early hypotheses about fission.
He ends up making the most convincing case I’ve ever heard for investigating “cold fusion”—now referred to as LENR, Low-Energy Nuclear Reactions, in an attempt to exorcise the field’s association with earlier failures.
Alex Telford offers another interesting look at methodology, but this time in biology—well, biology following an example from video games.
[Super Mario] Speedrunners contest world records by having either the best execution of the fastest known route, or by discovering and exploiting newer, faster, routes. In September, the world record 120-star run stood at 1 hour, 37 minutes, and 35 seconds — compared to the 20 or so hours it takes your average player. In his world record run, the speedrunner Weegee plays with sustained machine-like precision; it’s hard to believe that anyone could shave off the extra 36 seconds needed to break through the 1:37 barrier. By now, the standard 120-star route is so thoroughly optimised that the time advantage from most new routes is measured in fractions of a second. It seemed there was little left to discover in this 27-year-old-game.
That is, until September, when a new ‘holy grail’ trick was discovered that cuts about a minute off the 120 star run time. What I found fascinating about this discovery is not the specifics of the trick per se, but the process by which it was found…..
Orthogonal Jones was found by an algorithm called “scattershot.”
Scattershot is a sort of empirical machine gun; it uses brute force search with random inputs to find the shortest path to collect a star from a given starting point. By simulating the potential paths of thousands, or millions, of Marios, scattershot can discover new optimal routes through levels….
The problem with random search methods is that the longer a sequence needs to be, the greater the processing power and time required to explore all the options; random search is inefficient for long input sequences. Krithalith got around this by incorporating learning into the algorithm, and saving shorter optimal move sequences that can be used as staging points for random exploration. These short sequences can then be chained together to find the best overall long sequence.
This, Alex suggests, “demonstrates that brute empiricism may be the only feasible route to discovering the possibilities present in even simple systems.” He then applies this to scientific discovery.
In practice, this means that a way to approach automating science could be to collect voluminous statistics on system states (deeply characterizing both healthy and unhealthy model systems, say) then using scattershot-like guided randomness to find interventions that lead to the shortest path between those states. This approach creates its own function that can then be optimized. If you’re trying to understand a disease, finding the quickest path to reversing that disease and restoring a healthy state tells you a great deal about which factors are most important in the disease process.
It’s an intriguing suggestion for how computing power and artificial intelligence might be able to amplify human rationality.
I’ll be sharing more from the other fellows in the next few weeks.